Origin: Problem-Driven Convergence (2022)
- In 2022, the team began to take shape through deep involvement in a number of highly complex real-world scenarios—ranging from high-concurrency information systems to physiological signal monitoring. Despite their diverse application contexts, these scenarios shared a common set of scientific challenges: observed data were highly non-stationary, heavily corrupted by noise, and constrained by limited computational resources, necessitating robust and reliable identification of latent patterns under dynamic, evolving conditions.
- To address this shared challenge, the two researchers initiated regular discussions centered on the modeling difficulties encountered in their respective projects, gradually establishing a collaborative mechanism rooted in periodic, iterative scientific dialogue. By exchanging insights on key issues such as time-series structure modeling, anomaly detection thresholding, and lightweight feature extraction, they uncovered methodological clues applicable across disparate application settings.
Dialogue: Cross-Domain Synergy (2023)
- In 2023, as several team members joined international academic institutions such as Belarusian State University, the group began integrating perspectives from diverse scholarly traditions. Researchers from China and Belarus engaged in in-depth dialogue across key areas—including performance time series analysis, signal pattern recognition, and optimization strategy design—addressing the aforementioned core issues. This exchange was grounded in a deliberate recognition of methodological complementarity.
- Through this process, the team progressively established three interdependent research pillars and fostered a collaborative culture characterized by open discussions, shared codebases, and joint problem definition. As a result, the team has now developed a robust online seminar framework and a unified technical infrastructure to support ongoing collaboration.
Institutionalization: Dual-Track Output (2024)
- In 2024, the team gradually expanded in size and, for the first time, achieved systematic collaborative outputs. These results were published under multi-author co-authorship arrangements in journals such as CMC, JIM, SPIE BDCIA, and ICMIDA. At the technology transfer level, the outcomes spanned areas including database performance monitoring and multi-terminal surveillance, laying the groundwork for a dual-track output framework encompassing "theoretical methodologies—technical implementation—practical validation."
- As collaboration expanded and the complexity of research outcomes increased, the team’s previous informal discussion model began to face challenges in terms of coordination efficiency and reproducibility. To address these issues, the team formalized several collaborative agreements in 2024: implementing standardized, reproducible experimental protocols; establishing a version-controlled, open-source code repository as a shared public infrastructure; and introducing a cross-timezone, topic-based rotating seminar mechanism. These measures collectively fostered a research agenda shaped collaboratively from diverse perspectives, marking a pivotal transition of the organization from a consensus-driven approach toward a more streamlined, institutionally supported framework.
Network: Distributed Commons (2025)
- As of 2025, the collaborative group has evolved into a distributed research network spanning multiple countries and regions, including China, Belarus, Finland, and beyond. Members come from diverse disciplinary backgrounds—ranging from computer science and applied mathematics to biomedical engineering and industrial automation—and the collaboration model has expanded to encompass multi-node coordination.
- Building on this foundation, the team has progressively refined its core concept of "data pattern analysis and state inference for complex scenarios," transforming it from an initial consensus into a unified paradigm guiding multidisciplinary exploration. This milestone marks the substantive emergence of a robust, distributed research community.